What are you working on?

My work is in theoretical computer science. This fi eld is essentially where computer science and mathematics meet. However, I consider myself more of a mathematician than an engineer. Within that framework, my focus is on algorithms. Basically, algorithms are the mathematical study of computer programs. We analyze what computers can do. For instance, there are some problems that cannot be solved effi ciently. Other problems can’t be solved at all by any program. Understanding what computers can and can’t do is what we do as algorithmicists.

I work quite a bit with geometry . That means getting computers to solve geometry problems like paper folding, or robotic arm folding, or, ideally, protein folding. We also work a lot with data structures––that is, organizing data so you can search through it, or answer various questions about it, effi ciently. Google is a really big data structure. In this structure, you want to be able to search for keywords, fi nd documents that contain all of these keywords, and then rank them. We also work with graph algorithms and network algorithms. In this case, you answer basic questions about a network––for example, where to put the fire stations in a city in order to minimize the damage in a worst-case scenario, or determining the average time it takes for a fi re truck to get to a random location.

Those are my main fi elds of study, but there are other things I do just for fun. I’m interested in what I call recreational algorithms, which are named after a fi eld called recreational mathematics. Martin Gardner is more or less the father of recreational math and he’s been a big infl uence on my dad (Martin Demaine) and me. So, we just study questions for play. For instance, can a computer play a game like Tetris? It turns out that it can, but not very well. Also, how does one design an algorithm to make art, or sculpture, or architecture? This might not be quite as serious as other things, but to me everything has the same motivation. They are interesting problems and fun to look at.

[ ... ]
for full text see VISIONS - MIT Interviews book